Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/superradcompany/tool-cli/llms.txt

Use this file to discover all available pages before exploring further.

This guide walks you through the core tool-cli workflow: finding a tool in the registry, calling it from the terminal, and wiring it up to your AI host app.

​
Install tool-cli

curl -fsSL https://cli.tool.store | sh
After installation, verify it works: 
tool --version

​
Install your first tool

Search the registry and install a tool: 
tool search bash

tool install library/bash

​
Inspect what it exposes

tool info library/bash
This shows all the tools, prompts, and resources the server exposes — exactly what an MCP client will see when it connects. To see the full input and output schema for a specific method: 
tool info library/bash -m exec

​
Call a method

Call any tool method directly from the terminal — no client needed: 
tool call library/bash -m exec -p command="echo hello"
Method shorthand: MCP tools often use toolname__method naming. You can use . as shorthand: 
# These are equivalent:
tool call library/bash -m bash__exec -p command="ls"
tool call library/bash -m .exec    -p command="ls"

​
Register with your AI host

Add the tool to your AI host app so it appears in Claude, Cursor, VS Code, or any other supported client: 
tool host add claude-code library/bash
Or add all your installed tools at once: 
tool host add claude-code
tool-cli creates a backup of your host’s config file before making any changes, so your original config is always safe.

​
Next steps

Using tools

Search, preview, install, and manage MCP tools.

Build a tool

Scaffold a new MCP server and publish it to the registry.

Host integration

Register tools with all your AI host apps.

Command reference

Full reference for every command and flag.

Build docs developers (and LLMs) love

Get started for freeTalk to us